Seeing the invisible - from neutrons to photons
Lead Research Organisation:
Bangor University
Department Name: Sch of Computer Science & Electronic Eng
Abstract
A world where nuclear fusion helps meet humanity's energy needs is now within reach but there is still no way of "seeing" the operation of fusion reactors in real-time, presenting critical operational and safety risks. This project will lead to a disruptive new sensor technology enabling monitoring of the operation of fusion reactors in real-time, directly addressing this urgent need.
Nuclear fusion will become a commercial proposition in the next decade revolutionising energy generation to supply abundant, clean energy. Conditions for light nuclei to fuse are extreme: hot plasma is held at 150-200 Million C by powerful magnets. This is accompanied by emission of highly energetic fast neutrons with 14.1 MeV energy. Materials adjacent to fusion reactions must tolerate very high temperatures and damaging neutrons so developing sensors and sensor materials capable of measurements in such conditions are among the greatest challenges.
This project will directly address these urgent drivers by delivering an entirely new class of durable inorganic glass scintillators, which convert neutrons to detected photons. These will be capable, for the first time, of detecting fast 14.1MeV neutrons emitted from fusion reactions at high temperatures, enabling real-time insight into operation of fusion reactors, far advanced from current state-of-art. This is timely as UK fusion transitions from lab- to pilot- to commercial-scale (e.g. STEP) as the need for real-time, robust sensors capable of years of operation is urgent.
Measurement methods for neutron flux in high intensity areas are few and new approaches are needed for next generation tokamaks. Fission chambers and gas filled detectors are fragile and surveillance foils do not provide real-time information. No technology yet exists capable of doing what we are attempting. Our novel sensors will enable a step-change by providing operators real-time measurements in extreme environments, accelerating design processes and enabling more efficient and advanced control mechanisms, greatly enhancing safety. Inorganic glasses can be produced at scale and are tolerant to damaging neutron radiation and high temperatures. However, current inorganic glass sensors cannot reliably detect fast 14.1 MeV neutrons from nuclear fusion as there is little scintillation. Plastic and liquid scintillators (including organic glasses) are sensitive but have very low tolerances to high temperatures and radiation damage. Developmental diamond-based sensors are small (< 5 cm) and cannot be produced at scale.
Our new inorganic glasses capable of detecting fast neutrons will bring game-changing advances in neutron detection for fusion energy. The most exciting potential rewards of this high-risk project will be acceleration and enhancement of development, design, construction, and operational safety of commercial nuclear fusion power plants to be built in the UK and globally in the next decade.
Nuclear fusion will become a commercial proposition in the next decade revolutionising energy generation to supply abundant, clean energy. Conditions for light nuclei to fuse are extreme: hot plasma is held at 150-200 Million C by powerful magnets. This is accompanied by emission of highly energetic fast neutrons with 14.1 MeV energy. Materials adjacent to fusion reactions must tolerate very high temperatures and damaging neutrons so developing sensors and sensor materials capable of measurements in such conditions are among the greatest challenges.
This project will directly address these urgent drivers by delivering an entirely new class of durable inorganic glass scintillators, which convert neutrons to detected photons. These will be capable, for the first time, of detecting fast 14.1MeV neutrons emitted from fusion reactions at high temperatures, enabling real-time insight into operation of fusion reactors, far advanced from current state-of-art. This is timely as UK fusion transitions from lab- to pilot- to commercial-scale (e.g. STEP) as the need for real-time, robust sensors capable of years of operation is urgent.
Measurement methods for neutron flux in high intensity areas are few and new approaches are needed for next generation tokamaks. Fission chambers and gas filled detectors are fragile and surveillance foils do not provide real-time information. No technology yet exists capable of doing what we are attempting. Our novel sensors will enable a step-change by providing operators real-time measurements in extreme environments, accelerating design processes and enabling more efficient and advanced control mechanisms, greatly enhancing safety. Inorganic glasses can be produced at scale and are tolerant to damaging neutron radiation and high temperatures. However, current inorganic glass sensors cannot reliably detect fast 14.1 MeV neutrons from nuclear fusion as there is little scintillation. Plastic and liquid scintillators (including organic glasses) are sensitive but have very low tolerances to high temperatures and radiation damage. Developmental diamond-based sensors are small (< 5 cm) and cannot be produced at scale.
Our new inorganic glasses capable of detecting fast neutrons will bring game-changing advances in neutron detection for fusion energy. The most exciting potential rewards of this high-risk project will be acceleration and enhancement of development, design, construction, and operational safety of commercial nuclear fusion power plants to be built in the UK and globally in the next decade.
Organisations
- Bangor University (Lead Research Organisation)
- Glass Technology Services (Collaboration, Project Partner)
- STFC Laboratories (Collaboration)
- Culham Centre for Fusion Energy (Collaboration)
- UNIVERSITY OF BIRMINGHAM (Collaboration)
- National Nuclear Laboratory (Collaboration, Project Partner)
- Scintacor (Collaboration)
- University of Birmingham (Project Partner)
- CCFE/UKAEA (Project Partner)
- Science and Technology Facilities Council (Project Partner)
- Scintacor Ltd (Project Partner)
Publications

Ghardi EM
(2024)
First-principles study of lithium aluminosilicate glass scintillators.
in Physical chemistry chemical physics : PCCP
Description | Existing neutron detecting glasses were discovered in the 1960s. They are not, however, sensitive to the very fast neutrons produced during nuclear fusion. Fusion power may well be at an epoch defining point, as it transitions from a laboratory experiment to the practical source of abundant, clean power that it has always promised to become. This means that the need for resilient neutron sensors capable of helping to control fusion power stations is increasingly important. This project has been working to address this challenge, by harnessing the radiation tolerance of glasses and allowing them to be used to detect fast neutrons. A combined programme of materials design, materials synthesis and neutron testing has produced a list of new glass compositions that show promise in this application. At the time of writing the project is only weeks away from testing our new glasses with neutrons to see how they perform. Given the last major step forward in this field occurred in the 1960s, our work represents a significant step forward in the field of scintillating inorganic glasses. |
Exploitation Route | Durable glass neutron detectors are useful in a number of applications. A key property of glasses is that they can be made into a much wider range of sizes and shapes than traditional sensors. Single crystal scintillators have a maximum size in the centimetre range (and arrays of sensors rapidly become prohibitively expensive). Thin film sensors have similar problems whilst organic (plastic) materials can only be used at relatively low temperatures, limiting applications. Our glasses promise that they can be formed into large area detectors in shapes such as fibres and sheets, which opens up a large number of applications. Neutron detectors in fusion power stations Detectors in high energy experiments (e.g. the types of experiments conducted at CERN) Detectors in nuclear fission reactors Medical imaging Homeland security and border protection (detect illegal transport of nuclear materials) Geology, oil and gas prospecting |
Sectors | Aerospace Defence and Marine Digital/Communication/Information Technologies (including Software) Electronics Healthcare Government Democracy and Justice Pharmaceuticals and Medical Biotechnology Security and Diplomacy Transport |
Title | Machine learning guided glass discovery |
Description | Dr. Mehdi Ghardi developed a set of computational tools at Bangor University to accelerate the discovery of new glass compositions. The aim of this was to find glasses that scintillate when exposed to the fast neutrons emitted during nuclear fusion. He used an automated machine learning approach to identify areas of compositional interest that have been used to inform the experimental synthesis of new glasses. Ideally, rigorous physics based simulation codes would be used to predict the behaviour of candidate materials as part of these search algorithms. Whilst physically accurate, the codes we use (GEANT-IV) are computationally demanding meaning their run times are too long to be useful in this application. This is because the search algorithms tend to converge to solutions slowly and would and would require many thousands of simulation runs to complete, making them impractical. In order to address this, Dr. Ghardi used the physics based simulations to generate a dataset showing the response of important isotopes to incident radiation. These were used to train other machine learning models, which could be evaluated quickly, allowing their use as surrogates within the material discovery phase. Constraints were applied to the search to allow reasonable materials to be obtained. These including predictions of glass forming ability and glass transition temperature. By using this tool a database of promising glass compositions has been obtained. This has produced a ranked list of glasses that are new to the field of nuclear scintillators. These have been passed to the team at Sheffield Hallam University for experimental consideration. |
Type Of Material | Computer model/algorithm |
Year Produced | 2023 |
Provided To Others? | No |
Impact | The algorithm and resulting database has identified a number of glasses that have not previously been considered as neutron sensors. Consequently it has opened up new avenues of investigation. The application of machine learning has allowed a wide area compositional space to be surveyed efficiently. This has allowed experimental resources to be targeted more efficiently. The new compositions discovered may be patentable and work is proceeding in this direction. The outcome of this will determine our strategy to publish the results in the scientific literature. |
Description | Culham Centre for Fusion Energy |
Organisation | Culham Centre for Fusion Energy |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | The research team has arranged workshop sessions during which technical partners, including CCFE were included. |
Collaborator Contribution | CCFE has engaged in technical workshops involving all our industrial and national lab partners. Their attendance and the discussions that were had, have helped shape the direction of the research. The particular contribution of CCFE has been to highlight the neutron sensing methods currently used in the fusion experiments used at Culham. These have shown the limitations of current approaches and established a baseline which any sensor, developed from this project will need to exceed. |
Impact | Outputs are still pending. |
Start Year | 2022 |
Description | Glass Technology Services Ltd |
Organisation | Glass Technology Services |
Country | United Kingdom |
Sector | Private |
PI Contribution | Glass technology services have attended regular workshops involving the industrial and national lab partners involved with this project. Researchers on the project, in particular those involved with the experimental synthesis of glasses (Sheffield Hallam) enjoy a good working relationship with GTS. They have regular technical engagement. |
Collaborator Contribution | GTS have made useful contributions to workshop sessions held during the project. In particular they have provided useful technical advice on the practicalities of synthesising some of the more exotic glass compositions that have resulted from this research programme. |
Impact | Outputs are still pending. |
Start Year | 2022 |
Description | National Nuclear Laboratory (NNL) |
Organisation | National Nuclear Laboratory |
Country | United Kingdom |
Sector | Public |
PI Contribution | The research team has organised regular workshop sessions to inform and engage our national lab and industrial partners. NNL have regularly attended these events. |
Collaborator Contribution | NNL have provided useful comments on how the technology developed from the project may be applied within the nuclear industry (both fission and fusion). |
Impact | Outputs involving NNL are still pending. |
Start Year | 2022 |
Description | STFC Laboratories |
Organisation | STFC Laboratories |
Country | United Kingdom |
Sector | Public |
PI Contribution | The research team associated with this project have organised regular workshop sessions to report on progress and enable discussions with our national lab and industrial partners. STFC have been well represented at these sessions. |
Collaborator Contribution | STFC have provided valuable input during our workshop sessions. In particular they have helped form our strategy and planning around the irradiation experiments planned for the later stages of the project. During these experiments, glasses developed during our research will be exposed to neutrons and have their performance assessed. STFC have provided useful advice on the types of tests that should be conducted and the outputs that may be expected. |
Impact | Collaborative outputs are expected after material testing has taken place in March/April-2024. |
Start Year | 2022 |
Description | Scintacor |
Organisation | Scintacor |
Country | United Kingdom |
Sector | Private |
PI Contribution | Scintacor have been involved with discussion regarding new glass formulations suitable for sensing fast neutrons in nuclear fusion applications. This has been in the form of research meetings and inclusion in stakeholder workshops organised through this project. |
Collaborator Contribution | Scintacor have provided expertise, advice and materials samples to the project. The academic research team have visited their factory, where existing scintillating glasses are made. Reciprocal visits have also been made to the team at Sheffield-Hallam. Samples of their existing products have been provided against which new materials developed for this project will be benchmarked. |
Impact | Outputs still pending. |
Start Year | 2022 |
Description | University of Birmingham |
Organisation | University of Birmingham |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | The University of Birmingham are making their neutron irradiation facilities available for testing the scintillating glasses produced during this project. The project research team are currently engaged with Birmingham to design these experiments. |
Collaborator Contribution | The University of Birmingham are providing access to the neutron source to test our scintillating glasses. These experiments are due to take place in late March 2024. These are an important part of the work as they will provide a practical demonstration of the materials produced. The University of Birmingham have already provided useful input into the experimental design and will be providing access to their facilities as a valuable in-kind contribution to the project. |
Impact | Outputs will be realised in March 2024. |
Start Year | 2022 |
Title | Machine Learning Guided Glass Discovery Tools |
Description | The algorithms for glass scintillator discovery listed in the "Research Datasets, Databases & Models" section have been implemented in a set of software tools. These are primarily written in python, meaning they may easily be adapted to other material discovery missions. Currently the software is configured to find glass compositions likely to produce scintillation when exposed to fast neutrons. |
Type Of Technology | Physical Model/Kit |
Year Produced | 2023 |
Impact | The software has been used within the project to produce a database of candidate glasses that may be applied as scintillators for neutron detection within the fusion power stations of the future. |